Schema.org: definition, vocabulary and implementation
Updated on February 22, 2026
Quick definition
Schema.org is a standardised semantic vocabulary of structured data created in 2011 by Google, Bing, Yahoo and Yandex to allow search engines to better understand the context and meaning of web page content. Schema.org defines entity types (Product, Article, Event, Organisation) and their properties — it is the technical foundation that enables rich snippets in the SERP.
How it works
Schema.org was launched on 2 June 2011 to create a common vocabulary allowing webmasters to communicate the meaning of their content to search engines. Before Schema.org, several competing formats coexisted without standardisation.
Schema.org defines a hierarchy of types: at the top is `Thing`, from which derive `CreativeWork` (including Article, Recipe, Book, Movie), `Event`, `Organization`, `Person`, `Place`, `Product` and many others. Each type has specific properties — for example, the `Product` type exposes `name`, `description`, `price`, `aggregateRating`, `availability`.
The vocabulary is implemented via three formats:
- JSON-LD (recommended by Google): script separate from HTML, easy to maintain
- Microdata: inline HTML attributes, still supported
- RDFa: supported but rarely used
Concrete example: a restaurant implements the `LocalBusiness` type with `name`, `address`, `telephone`, `openingHours`, `priceRange` and `aggregateRating`. Google can then display a rich snippet with the rating, opening hours and price range directly in local SERPs.
Why it matters
Schema.org is the bridge between your content and the semantic understanding of search engines. Without structured data, Google has to infer the type and properties of your content from text alone.
With Schema.org, you communicate this information directly, opening the door to rich snippets, knowledge panels, carousel results and voice assistants.
- In 2026, with the rise of generative search (Google's AI Overviews), structured data becomes even more important
- E-commerce sites with `Product` + `aggregateRating` benefit from a significant CTR advantage
- `FAQPage` and `HowTo` can multiply the visual space occupied in the SERPs
How to improve or use it
- 1Identify the most precise Schema.org type for your main content (Article, Product, FAQPage, LocalBusiness, Event).
- 2Implement the markup in JSON-LD in the `<head>` or `<body>` section of your pages.
- 3Include all required and recommended properties for your type.
- 4Validate with the Google Rich Results Test.
- 5Monitor errors in Google Search Console's Schema.org report.
- 6Prioritise types that trigger visual rich snippets (Product, Review, Recipe, FAQPage).
With Sublim
Sublim lets you measure the impact of your Schema.org implementations on organic traffic: by comparing CTR before and after activating rich snippets on your key pages, you concretely validate the return on investment of each type of structured data. All measurements are performed cookieless, in compliance with GDPR, with European hosting.
Frequently asked questions
Are Schema.org and JSON-LD the same thing?
No, Schema.org is the vocabulary (the types and properties: Product, Article, aggregateRating, etc.), while JSON-LD is the encoding format (the JavaScript syntax used to write that vocabulary in a web page). We say that we 'implement Schema.org in JSON-LD'. Other formats can also implement Schema.org: Microdata and RDFa, but JSON-LD is recommended by Google.
Is Schema.org useful for small sites?
Yes, Schema.org is beneficial regardless of the site's size. For a small local site (restaurant, craftsman, service provider), implementing LocalBusiness with opening hours, address and reviews can trigger a rich snippet in local results and significantly increase CTR. For a blog, Article and FAQPage are accessible types that can improve visibility without much technical complexity.
How many Schema.org types can be used on a single page?
A single page can contain multiple Schema.org types — for example, an Article containing a Review and an FAQPage at the bottom. However, Google recommends marking up only the page's main and relevant content. Marking up content unrepresentative of the page or using inappropriate types can lead to rejection of rich-snippet eligibility.
Related terms
Structured data is a standardised semantic markup added to a web page'…
A rich snippet is an enhanced search result displayed in the [SERP](se…
The SERP (Search Engine Results Page) is the page displayed by a searc…
SEO (Search Engine Optimization) is the set of practices aimed at impr…